I think the biggest argument for keeping constraints in the model is that it could be used in the future to help Cayenne determine better orders of operation.
mrg On Mon, Oct 10, 2011 at 10:11 PM, Aristedes Maniatis <[email protected]> wrote: > On Mon Oct 10 20:32:13 2011, Durchholz, Joachim wrote: >> >> Just giving feedback from my user perspective: >> >>> CayenneModeler is not a DB admin tool, so DB concepts not directly >>> related to ORM are generally ignored. >> >> I find that reasonable. >> >>> UNIQUE indexes are somewhat on the border and I even believe we have a >>> feature request in Jira somewhere... >> >> This I cannot agree with, I do think unique indexes are solidly on the ORM >> side. >> The issue is relationships defined using unique keys. These tend to come >> from two sources: >> - Legacy databases that I need to interface with. >> - People with direct manual SQL access and a say on the data model. For >> them, having a UK with well-chosen content can make the difference between >> needing a join or not, so they do have a point. >> >> Regards, >> Jo > > I agree that unique indices are important to the model, just like validating > that a varchar is no more than (say) 12 characters. Personally I'd like to > see all indices in the Cayenne model if only because it means we have one > place to store the entire db schema definition, but I do understand why this > should be kept separate to the Cayenne runtime. > > -- > --------------------------> > Aristedes Maniatis > GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A >
